%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
INSTRUCTIONS TO REPLICATE THE ANALYSIS IN THE PAPER
"Household Portfolios and Implicit Risk Preference"
by Alessandro Bucciol and Raffaele Miniaci
Review of Economics and Statistics, 93(4), 1235-1250.

written by Alessandro Bucciol
University of Verona
Email: alessandro.bucciol@univr.it
June 26 2010
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%



The content of the zip archive is divided in two main folders:

- "\code": contains the Matlab codes used to estimate the implicit risk tolerance coefficient, and the Stata codes used to perform the analysis shown in the paper.
- "\data": contains the original datasets and the Stata code used to create portfolios from the SCF.

Each folder contains a text file "_README.txt" with further instructions.



The "\data" folder contains the raw data on time series asset prices (in "\data\series" and household investments (in "\data\portfolios"), as well as instructions and the code to derive the asset returns and household portfolios used as inputs in the analysis. The folders also contain the datasets resulting from this calculation, so one does not need to run the code and create the input data again.
A further folder, "\data\estimates", contains all the final datasets with the implicit risk tolerance estimates. These are created with the computer codes in the "\code" folder.


The "\code" folder contains the Matlab and Stata programs used in the analysis shown in the paper.

Run the Matlab program "MASTER" in the "\Matlab" folder to estimate the risk tolerance implicit from the observed portfolio composition. Final datasets are saved in the "data\estimates" folder as Matlab file ("macro.mat", for the aggregate portfolios) and text files ("micro_1.txt"-"micro_5.txt", for each imputation of the household portfolios).

Run the Stata program "MASTER" in the "\Stata" folder to perform the analyses on the risk tolerance coefficient shown in the paper. Final datasets are saved in the "data\estimates" folder as Stata files ("micro_1.dta"-"micro_5.dta", for each imputation of the household portfolios).
